• Share
  • Email
  • Embed
  • Like
  • Private Content
Eyeblaster: Trends in Conversions
 

Eyeblaster: Trends in Conversions

on

  • 4,839 views

Excelente white paper, lo que más me gusto es que muestran las tendencias del CTR por zona mundial, México & Latam entran en " North America "

Excelente white paper, lo que más me gusto es que muestran las tendencias del CTR por zona mundial, México & Latam entran en " North America "

Gracias Lazaro por el archivo

Statistics

Views

Total Views
4,839
Views on SlideShare
4,763
Embed Views
76

Actions

Likes
1
Downloads
138
Comments
0

1 Embed 76

http://neuronadigital.ning.com 76

Accessibility

Categories

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

    Eyeblaster: Trends in Conversions Eyeblaster: Trends in Conversions Document Transcript

    • Eyeblaster Analytics Bulletin Issue 3 | Trends in Conversions March 2009
    • Background Summary of findings In a time of recession, • Click-thru Rate vs. Conversion Rate: Worldwide (WW) ’08 CTR is 0.14% accountability of online while WW’08 Conversion Rate stands at advertising becomes more 0.32% - more than double. critical than ever. Whether you • Correlation between Format Type and Conversion Rate: Rich media ad formats are pursuing a direct response increase the tendency of users to convert by 2.5x. initiative or looking to achieve • Impact of Data Capture on Conversion measurable branding objectives, Rate: Lowering the number of phases the the focus is on the process of user needs to go through by bringing the conversion mechanism into the banner. `conversion’: that moment when You are 5x more likely to convert in a an individual exposed to an ad banner than you are on a site. performs a desired action.1 • CTR vs. Data Capture: Comparisons across unique data reveals users are This issue of the Analytics nearly 8x more likely to convert in a banner than click on it and about 34% of Bulletin focuses on the role unique users who interact will fill in form details within a banner. of conversions. Given the • Automatic Optimization and its impact accountability and measurability on Conversion Rate: Conversion rate of online media, what measures of campaigns using an optimization algorithm is significantly higher (1.25x) can we take in order to maximize than the average conversion rate of conversions at zero or minimum rich media and practically double the global average conversion rate for all ad cost? Eyeblaster conversion formats. tags give the ability to track • Search as a conversion metric: One can measure an increase in search activity user’s activity on the advertiser’s as a result of display advertising that is site after viewing or clicking on sustained for a sizeable amount of time after the display campaign has finished. an ad, but this is only one of a • Behavioral sequencing: Consumers myriad of methods that can be exposed to targeted messages in deployed to ensure a better yield sequence are more likely to move through a conversion sequence. of success. Pg 193 – Marketing Sherpa’s 2008 1 Pg 193 - Marketing Sherpa's 2008
    • Table of contents Research Methodology ............................................................................. 4 Results in Detail ........................................................................................ 4 Post-Click vs. Post-Impression ................................................................. 5 Improving the conversion process with Data Capture ................................ 6 Automatic Optimization mechanism and its impact on conversion rate ..... 10 Search within the conversion process .......................................................11 Benchmarks Appendix 1: north America .................................................................... 15 Appendix 2: Europe ................................................................................ 16 Appendix 3: Spain .................................................................................. 17 Appendix 4: APAC .................................................................................. 18 Appendix 5: Taiwan ................................................................................. 19 3 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Research Methodology Results in detail This study is based on data of more than 2,000 Click-thru Rate (CTR) vs. Conversion Rate campaigns representing 29 billion impressions collected CTR is a valuable metric to measure the amount of users between Jan ’08 and Dec ’08 covering all sectors. who wish to visit the advertiser’s site, but this in itself is not These campaigns represent advertisers using Eyeblaster a true conversion. The common perception is that CTR conversion tags. Conversion rates were based on indicates campaign success; the following analysis shows total conversions with no differentiation of the type. that low or high CTR does not necessarily correlate with All calculations were performed on daily delivery data the campaign’s overall objectives. Conversion tags placed aggregated to the ad level. on the advertiser’s web site on the other hand, track actual user activity that the advertiser wishes to track. Thinking about conversions… In observing the CTR metric alone, we expect all consumers to react immediately and overlook consumers who converted after being exposed to – and possibly interacting with – the ad. In order to see the full picture we should sum both types It could be argued that of conversions. a true conversion would be where a consumer CTR vs. Conversion Rate has made a monetary 0.35% transaction directly with the 0.32% advertiser, i.e. purchased 0.30% something. The focus here 0.25% is on conversions where .5 x2 0.20% the conversion means 0.14% ‘showing intent’ – such 0.15% as through data capture 0.10% – indicating a consumer’s 0.05% desire to purchase. 0.00% It is worth noting that a total conversion CTR Total Conversion Rate could happen directly on an advertiser’s site Eyeblaster Jan.2009 but possibly more so on a reseller’s site or in some other place, such as offline or in-store. The true effects of digital advertising against Figure 1 Source: Eyeblaster Research 2009 this overall conversion process (included within or in isolation from other media) are In looking at the overall lift in the conversion rate, it not considered here due to current tracking indicates that a higher percentage of consumers take in methods. For the purpose of this document, the advertising message, then require time to digest the we shall limit conversions to those indicating information, possibly clarify it from other sources, before intent within the ad itself or direct on an finally being in a position to actually convert. advertiser’s site, appreciating this in itself is only showing a slither of the actual overall Conversion rate provides more accurate and precise conversion rate and ROI. measurement of the campaign’s success and global data reveals that by focusing on CTR alone, we are missing 2.5x the actual data an advertiser requires to correctly analyze the overall consumer journey. 4 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Post-Click vs. Post-Impression Standard advertising plays an important role in enhancing reach and frequency of the entire media mix (Rich, Standard, Search) so while not having as strong a recall Correlation between Format Type and and conversion rate, it acts as a reminder for those already Conversion Rate with awareness, and as a reach extender. We shall look at ways to enhance the standard banner effectiveness later. With pressure mounting on trying to ascertain the most cost-effective online advertising, it would seem that high Another way to look at this conversion process is to try to volume standard display advertising would naturally be a understand when the conversion ultimately takes place, first choice. however, when we look at the true conversion i.e. is the advertising message creating an immediate rates for the 3 most popular ad format types online, we or a time delayed conversion? Previously we have seen clearly distinguish a higher rate for rich media advertising. that rich media is creating stand-out to ensure brand recall later, but is this a trend we should expect when we Conversion Rate by Formats consider all display advertising? Let’s compare the following two scenarios: • See-Click-Convert (post click) 0.60% • See-Delay-Convert (post impression) 0.50% 0.47% 0.50% Conversion Rate 0.40% 0.16% 0.30% 0.14% 0.20% 0.14% 0.20% 0.12% 0.10% 5 .7 0.10% x1 0.00% 0.08% STD Banner Polite Banner Expandable 0.08% Banner 0.06% Eyeblaster Jan.2009 Figure 2 Source: Eyeblaster Research 2009 0.04% 0.02% Both popular rich media formats are showing 2.5x 0.00% Post - Click Post - Impression uplift in the ability to generate conversions over standard Figure 3 Source: Eyeblaster Research 2009 display inventory. The results indeed show that there is a sizeable uplift in the number of consumers who do not necessarily respond The results reveal that rich media is doing two things and convert immediately to an advertising message, here. First, it is attracting attention of the viewer against and this factor needs to be explored further in trying to site content, and thereby ensuring greater stand out understand the consumer’s mindset. from site editorial. Second, by allowing a consumer to tangibly touch and interact with the advertising message, the consumer experience is enhanced through physical 1.75x conversions happen involvement, with a resulting positive effect for the advertiser, shown in actual conversions. after a time delay 5 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Improving the conversion process with Data Capture Whether you go for standard We all know that in the “viewer to converter” funnel there is a drop off at every step. Therefore, we should aim to or rich display advertising, minimize the funnel steps. In general, data capture is the ability to collect information submitted by the user. By not including data capture placing data capture functionality into the advert itself; we are addressing a type of conversion that occurs on within any DR campaign is at the web site where the advert is placed, instead of the advertiser’s web site. This feature can be implemented in least halving your overall ROI various ways: one of the most common is allowing users to request additional advertiser information from within the effectiveness. ad. It can also be used for polling users’ opinion, collecting email addresses, cell phone numbers, or any other details, including full transactional details. “Viewer to Converter” Funnel On-Site conversion in-Banner Conversion Rich Regular banner flow (Rich Media banner flow or Standard) Viewing and/or interacting Viewing and/or interacting with the ad with the ad Clicking on the advert Arriving at the Details Submission web page Sign in-start page Details Confirmation Sign in - Confirmation page Figure 4 Source: Eyeblaster Research 2009 6 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • You are 5x more likely to Conversion Rate (on-Site vs. in-Banner) convert in a banner than convert on a site as a result of 1.20% 1.06% clicking on a standard banner 1.00% x5 0.80% Case Study Impact drives Conversions 0.60% health and Beauty campaign 0.51% 0.40% Let’s try to clarify the power of data capture by 0.20% referring to the following case study. 0.20% In a health and Beauty campaign we monitored the amount of email addresses submitted on 0.00% a daily basis during the campaign’s lifetime, 22 Standard Rich Banner Data Capture days. (Over 3 million impressions were served on Banner Yahoo, nine.MSN and news.com.) Figure 5 Source: Eyeblaster Research 2009 in-Banner vs. on-Site Conversions When comparing the data capture within a banner 60 (1.06%) with that of on-site (0.20% and 0.51%) we Conversion or Form Submission see that by removing a number of required steps, the 50 conversion rate is 5x that of standard display and 40 double that of rich media as a result! 30 By simplifying the process that the consumer is required to follow, the advertiser achieves a more desirable result. 20 however, it is important to note that the level of interactivity 10 within Rich Media is nOT a bearing on the conversion 0 result here; it is purely about making things easier for the 1234 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 consumer to connect and convert in general. users want Days from campaign’s start date to connect with a brand where they are. Form Submissions Confirmation Conversions When we consider that standard advertising does not Eyeblaster Jan.2009 have the ability to measure anything beyond a click, Figure 6 Source: Eyeblaster Research 2009 the need for inclusion of a data capture facility across networks is startlingly obvious. not only is this challenging the notions of cost-effectiveness of standard media buys The amount of emails submitted at driving conversions, but also what we have come to from the banner itself (data define as Direct Response advertising online. It would serve all standard advertising well to move towards the capture) was significantly higher inclusion of simple response based functionality. throughout the campaign’s lifetime in comparison to emails submitted from the advertiser’s site. 7 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Further challenging the notion of CTR as an Personalizing the “Viewer to Converter” effective conversion metric Funnel With CTR being the de-facto measurement of online DR Impressions and CTR are non-personal attempts at trying campaigns, how does this need to be addressed given the to understand what consumers actually do online. In wealth of data on consumer behaviour discovered over the breaking down the “viewer to converter” funnel for those last 10 years about rich media advertising? who convert in banner, we need to try and understand how many times does a consumer get exposed to an advert, how many touch the advert, for how long do they interact Unique Clicking Users vs. Form Submission with it and what is the resulting conversion pattern? Therefore we see a funnel of total impressions ➔ unique impressions ➔ unique interacting users ➔ number of 1.20% conversions (which we will determine as form submission 1.06% and assuming that person will not submit more than one 1.00% form and thus can be related to as a unique number.) • See-Interact-Convert (in-Banner) 0.80% 0.60% 77 x7. 0.40% 0.20% 0.14% 0.00% Click Through Rate In - Banner Conversion Rate Figure 7 Source: Eyeblaster Research 2009 • unique viewers’ rate – unique out of total impressions using global data, we can see that the inclusion of data capture within a banner is just shy of 8x more effective at • unique Interacting users - unique interacting users out addressing conversions than CTR. It begs the question, of unique impressions why don’t all DR campaigns have basic data capture • Average Dwell Time – number of seconds a viewer functionality built-in as standard? actively is engaged with an advert Consumers are nearly 8x more • Converting users – form submissions out of unique interacting users likely to fill in a form in a banner 34% of unique users who then click on it interact will fill in form details within a banner What we see from the data is the viewer is being exposed to the ad at least twice (42.79%). 7.40% of these then go on to interact with the ad, and do so for well over a minute (78.08 secs). 33.56% of those who interact go on to convert within the banner. 8 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Communication and conversion Case Study People love to talk. Whether it is product feedback or recommending to a friend, there are a variety of methods e-Commerce transactional to start to penetrate the consumer conversation to ensure higher conversion rates. Widgets within social networks banner are one such method. But what is more interesting is that due to the poised typing position people have within How far can we push conversions communication channels, such as email and Messenger, there is often more of a willingness to fill in data capture in-Banner? forms within these channels. A leading European mobile operator wanted to attract new subscribers to their network. The Health Campaign concept of data capture was taken to the next CTR vs. Data Capture stage as they wanted to get much more personal 0.70% attributes from the interested user as part of 0.60% entering into a contract with the operator. In 0.60% this case not just address details, but additional 0.50% qualifiers such as passport or ID card number as well as bank account details. The incentive 0.40% to do so was an offer for a free mobile handset. .06 x4 0.30% The placement of a creative on a leading portal seemed to offer assurance to the 0.20% 0.10% interested consumer and combined with the 0.10% incentive was enough to inspire confidence of 500 consumers who went on to give full 0.00% personal details within the advert! This level Click Through Rate In - Banner of success was well-beyond the expectations of Submission the advertiser, but it goes to prove the willingness of modern consumers – and the need to not Figure 8 Source: Eyeblaster Research 2009 overlook data capture within creative executions. Taking standard display advertising to the next level: A health and Beauty Campaign that utilized a standard display advert within a communication channel found that the CTR was under performing. however, the inclusion of a simple data-capture form that opened in an expansion panel upon clicking on a button within the creative saw a quadruple rise in in-banner submissions over the CTR, incentivized by the chance to win a pair of sunglasses. Inclusion of a simple data- capture form… saw a quadruple rise in in-banner submissions over the CTR This is a trend that is seen in Automotive, Travel and health who often utilize data capture as part of their campaign objectives. 9 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Automatic Optimization mechanism and its Case Study impact on conversion rate not only can you minimize the number of steps in the Data Capture as the “viewer to converter” funnel, there is an ability to monitor in desired call to action real-time the effects of one creative against another during the campaign. Eyeblaster’s Automatic Optimization mechanism uses Masterfoods hoped to compel interested an algorithm to find the ads with the highest success consumers to sample a new type of chocolate rate based on the metric defined, which can often vary without the need to click through to a mini-site. upon placement of media by section or publisher. The They created a contest to drive interest and case study below demonstrates the results of such engagement and a series of ads was developed a methodology deployed and the overall campaign performance based on conversions. around the idea of “Mix It up”. The prize was the chance to win a 3 month’s supply of chocolates Comparisons against similar in exchange for personal contact details. The executions were adapted to suit the publisher campaigns - one with environment; e.g. high impact on the homepage, with the ability to play a game prior to entry optimization and one without, into the competition – but data-capture alone reveals a considerable change within communication channels. All creatives were designed to remove steps to conversion in the ability to control a by immediately presenting the competition entry form on roll-over which was the focus, instead positive shift in ROI. of muddying the call to action with click-thru. It worked. In a single day 35,000 people entered details indicating they wanted to From data collected across one Automotive advertiser taste Mars chocolates! An incredible indication over a six month period, the conversion rate of campaigns to a brand as to audience favorability and a true using an optimization algorithm (0.63%) is significantly higher (1.25x) than the average conversion rate of rich demonstration of online DR at work. media ads (0.51%). In fact, it is practically double the average conversion rate for WW ’08 – 0.32%. The inclusion of an incentive, To understand this more fully, let’s delve inside a campaign that implemented real-time optimization. such as a giveaway or access to exclusive content, will positively 0.70% impact the data capture rates. 0.60% Automotive Advertiser (0.63%) 0.50% x1.25 Rich Media 0.40% (0.51%) x1.97 0.30% 0.20% WW ’08 (0.32%) 0.10% 0.00% Conversion Rate Figure 9 Source: Eyeblaster Research 2009 10 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Search within the conversion process Case Study As we have seen, not all display advertising messages can be tracked either by an immediate response within the A campaign using the banner, but neither can they be determined by a post-view arrival at an advertiser’s site. Automatic Optimization mechanism Influence of Online Display on Clicks of Search Campaigns Creative A is tested against Creative B in the same automotive campaign. All factors such as placement, ad format and size remain consistent, 2000 600 just the creative messaging was different – there 1800 were 5 messages per placement. Looking at the 500 1600 final conversion rate as the ultimate point to weigh against – irrespective of click-thru or interaction Impressions (in 1000s) 1400 400 rate – we are able to control a significant uplift in 1200 conversions. Clicks 1000 300 800 Conversion Rate with Automatic 200 Optimization 600 400 100 200 25.00% Weeks 0 0 1 2 3 4 5 6 7 8 20.00% Impressions of display campaigns Clicks of search campaigns 15.00% Figure 11 Source: Eyeblaster Research 2009 10.00% In some scenarios, after a consumer is exposed to a message, they wish to perform an independent evaluation, 5.00% which ultimately may even lead to a purchase offline or at a reseller’s site. Think about your recent purchase behavior – did you review information on a comparison site first? 0.00% And was that purchase ultimately made directly on the 1 3 5 7 9 11 13 15 17 19 21 23 25 27 advertisers site – if online at all? Therefore we need to look Conv Rate of a Regular f light Conv Rate of an Optimized f light at cross-channel activity in order to ascertain a deeper understanding of overall conversion effectiveness beyond Figure 10 Source: Eyeblaster Research 2009 in-banner or on-site conversions. In this example, we find that initially search activity is fairly The non-optimized flight was performing better dormant. Once the display activity has begun, there is a initially, but once the technology automatically followed lift in the search activity. What is important to note adjusted the weighting of the creative in the is that often there is a delay between the display exposure optimized flight, you can see the uplift in the trend. and the performing of the search, as a consumer finds their own time to do their research. When looking at the final data, we find that the optimized creative over time lifts the overall Optimizing creatives lifts conversion rate by 0.82% for the non-optimized message to create impressive spikes of 23.42% – the overall conversion rate and totally smashes the average global conversion rate of 0.32%. 11 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • 4500 4500 4000 4000 Thomas McIlheran, Senior Media Manager 3500 3500 says: “The Eyeblaster Channel Connect Impressions (in 1000s) Reporting system is the most progressive 3000 3000 reporting tool for monitoring the joint impact of 2500 2500 Clicks search and display advertising. CC4S allows 2000 2000 advertisers and marketers to put numbers around exactly how well these online tactics 1500 1500 work together as opposed to how they work 1000 1000 against or in spite of one another.” 500 500 0 0 03/10/ 17/10/ 31/10/ 14/11/ 28/11/ 12/12/ 12/12/ 09/01/ “Using Channel Connect for Search and 2008 2008 2008 2008 2008 2008 2008 2009 Display, Mindshare was able to pinpoint crucial Impressions of display campaigns Clicks of search campaigns campaign data and draw important insights Figure 12 Source: Eyeblaster Research 2009 about the interaction of our search and display ads. In the end, it provided us with a more comprehensive overview of user behavior”, The tailing-off effect of the search activity is maintained at a much slower decline than the display advertising, as Harry Case, Director of Media Analytics the emotional connections a consumer has made as a and Technology, Mindshare. result of the ad exposure continue to play on their minds for perhaps the next couple of weeks. It could also be fair to say that the multiple exposures to display messaging during this time are having a positive impact in the desire to search, and this higher frequency ensures a greatly drawn-out time span of searches – long after the effects Targeted messaging with behavioral can be seen within traditional display advertising metrics, sequencing i.e. clicks have declined, but searches are maintained. This inherently challenges traditional notions of how The ability to adaptively sequence creative messaging can to optimize display advertising frequency, based be done using various tracking methodologies utilizing upon display advert CTR. cookies and Flash Shared Objects. This has distinct uses to ensure relevance to the viewer. There is a delay between the First, you can ascertain if a user has previously been to the advertiser’s site, and begin to delve where on the site and display exposure and the the profile of the person based upon established criteria. This enables knowledge of the person exposed to the performance of the search - creative to be ascertained prior to being shown a creative and is known in Eyeblaster as DéjàVu. potentially by as much as 3 Second, there is the ability to learn what creatives have weeks! been previously shown and/or interacted with and adapt the creative messaging accordingly. This is known as behavioral sequencing. From campaigns we have looked at already, we have Finally, there is also the ability to read in host-site found a positive shift towards brand/product name content of where the ad is sitting and adapt the creative searches as a result of the display activity, proving recall messaging to the surrounding page content. Called effect of display advertising. We have also seen uplift in Mash-up technology, this is another way to increase search activity as a result of display impressions. We shall advertising relevance. look forward to sharing such results in future bulletins. 12 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Case Study Sony Ericsson k550i campaign Sony Ericsson ran a campaign across CEEMEA, which adapted differently depending on how a user had interacted – and where – on site or within a banner. The creative was also changed each week to try and minimize ad fatigue through over exposure. http://creativezone.eyeblaster.com/#ItemName=SE%20K550 Total Data Impressions Clicks CTR Stage 1 Stage 2 Upload Send Custom Week 1 20,827,665 38,825 0.19% 1,704 249 25 1,978 Week 2 33,928,483 32,264 0.10% 12,416 85 1,058 200 13,759 Week 3 35,917,203 32,233 0.09% 20,245 423 1,600 280 22,548 Week 4 36,036,434 33,369 0.09% 17,978 111 1,827 357 20,273 Total 126,709,785 136,691 0.12% 52,343 619 4,734 862 58,558 Figure 13 Source: Eyeblaster Research 2009 Those exposed to a more The results demonstrated that as the creative fatigue happened with each week the campaign targeted message… were ran, even with changing the creative per week, more likely to move through CTR steadily declined due to over-saturation. the conversion sequence Stages 1 and 2 represent a targeted message showing a personal message for this week’s competition (stage 1) or if the user had already participated, a message advertising the coming week’s competition (stage 2) up to the final week 4. The data revealed that those exposed to a more targeted message through prior involvement, either on site or within the banner, were more likely to move through the conversion sequence; in this case uploading photographs alongside personal data and as a result, actively recommend the campaign to their friends with a forward-to-friend mechanism built into the ad unit. This is one of the most exciting findings of how activity on-site and in-banner can work together to engage consumers and lead them through a conversion cycle, as well as engage brand advocacy to their friends. 13 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Conclusion The potential to deliver better The advent of being able to link data across-channels is a revolutionary breakthrough for the advertising community ROI for clients is easily within and heralds the start of a completely new way of being able to effectively and accurately strategize media grasp for agencies and their and reduce overall marketing dollar wastage. We look forward, in future Analytics Bulletins, to show how this is clients progressing. We have shown how there are multiple ways to utilize conversions and gain higher ROI for advertisers. As with most software, between 5-10% of possible functionality is currently being utilized within most campaigns and thereby the potential to develop more cost-effective advertising is being lost through lack of knowledge of what is possible with modern day ad serving. As this functionality is generally included within the ad serving costs, coupled with a little creative strategy, the potential to deliver better ROI for clients is easily within the grasp for most agencies and their clients. Ask your Eyeblaster representative about how you can harness the potential of Eyeblaster’s Ad Campaign Manager to deliver greater ROI for your next client objective. 14 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Appendix 1 Benchmark for North America Q4 ’07 - Q3 ’08 Performance Metrics (Format and Verticals) Basic Metrics Video Metrics Expandable Metrics Interaction Impressions Avg. Avg. Video 50% Fully Total Average Started with any Panel Expansion IR CTR Duration Played Played Expansion Duration Rate Expansion Duration (Seconds) Rate Rate Rate (Seconds) Rate (Seconds) Standard Banner — — 0.09% — — — — — — — Rich Media Commercial Break 13% 2.73 2.61% 5.46 72% 10% 3% — — — Expandable Banner 11% 3.87 0.30% 23.62 15% 79% 69% 11% 7% 51.46 Floating Ad 3% 2.78 3.15% 6.64 48% 50% 34% — — — Formats Floating Ad With 5% 3.00 2.80% — — — — — — — Reminder Floating Expandable 10% 3.26 1.52% 5.26 62% 83% 70% 72% 5% 23.70 In Game 13% — 4.67% 16.60 94% 71% 47% — — — Polite Banner 5% 3.36 0.17% 19.76 53% 68% 52% — — — Push Down Banner 4% 2.29 0.28% — — — — 35% 3% 18.80 Video Strip 17% 4.02 0.17% 17.26 48% 59% 31% 46% 6% 70.23 Interaction Impressions Avg. CTR CTR Avg. Video 50% Fully Total Average Started with any Panel Expansion IR (Rich (Standard Duration Played Played Expansion Duration Rate Expansion Duration Media) Media) (Seconds) Rate Rate Rate (Seconds) Rate (Seconds) Apparel 12% 3.33 0.30% 0.24% 15.04 47% 74% 36% 20% 7% 30.41 Auto 9% 3.35 0.23% 0.11% 19.63 16% 76% 64% 17% 9% 31.20 B2B 6% 3.11 0.48% 0.17% — — — — 12% 9% 65.23 Careers 10% 3.45 0.30% 0.09% — — — — 24% 16% 53.43 Consumer Packaged 7% 3.63 0.35% 0.11% 24.90 25% 54% 33% 9% 6% 48.21 Goods Corporate 1% 3.09 0.09% 0.05% 22.05 7% 47% 31% 12% 8% 78.68 Electronics 8% 3.71 0.24% 0.12% 19.63 20% 66% 71% 15% 11% 42.02 Entertainment 11% 3.60 0.26% 0.13% 15.39 13% 61% 46% 9% 5% 49.49 Financial 5% 3.35 0.15% 0.07% 13.08 56% 76% 60% 6% 4% 46.69 Verticals Gaming 7% 4.35 0.63% 0.08% 59.46 17% 49% 35% 22% 8% 61.49 Government/utilities 5% 3.18 0.17% 0.07% 24.48 62% 78% 66% 16% 4% 26.07 health/Beauty 13% 3.29 0.24% 0.17% 30.80 8% 63% 41% 19% 13% 61.72 Medical 6% 3.07 0.14% 0.10% 21.88 38% 72% 58% 7% 5% 36.49 news/Media 10% 2.64 0.27% 0.07% — — — — 14% 9% 22.58 Restaurant 1% 2.94 0.13% 0.09% 16.04 50% 39% 27% 3% 2% 33.43 Retail 11% 3.09 0.29% 0.07% 23.48 11% 40% 25% 16% 9% 28.08 Services 8% 3.08 0.38% 0.05% 23.23 56% 77% 63% 33% 21% 70.24 Sports 7% 2.98 0.28% 0.08% 11.37 55% 68% 52% 7% 4% 44.50 Tech/Internet 4% 3.59 0.17% 0.14% 14.82 51% 77% 65% 6% 4% 88.84 Telecom 7% 3.28 0.18% 0.09% 23.55 34% 81% 62% 15% 10% 48.35 Travel 4% 3.80 0.21% 0.08% 40.20 59% 78% 61% 13% 5% 31.28 north America: including uS and Canada. The benchmark is for a rolling year, last updated Q4 2008. Interaction Rate (IR) user Initiated Interactions' divided by 'Served Impressions' Click-through Rate (CTR) Frequency of Click-throughs as a percentage of served impressions. 'Clicks' divided by 'Impressions'. Avg. Video Duration The average duration the video was played, including user and auto initiated videos (in seconds). Started Rate The number of times the video started out of video's served Impressions 50% Played Rate The number of videos that played over 50% of their total video length divided by 'Video Started'. Fully Played Rate The number of videos that were fully played divided by video that started Total Expansion Rate Total expansions divided by served impressions. Including Auto-initiated expansions Impressions with any Panel Expansion Rate The number of impressions with at least one panel expansion. Avg. Expansion Duration The average time a panel was expanded, including user and auto initiated expansions Rich Media Ads All Eyeblaster's format excluding standard banner, wallpaper and window ads — not Available - either this measurement is not applicable to the format or there was not enough data to be statistically relevant 15 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Appendix 2 Benchmark for Europe Q4 ’07 - Q3 ’08 Performance Metrics (Format and Verticals) Basic Metrics Video Metrics Expandable Metrics Interaction Impressions Avg. Avg. Video 50% Fully Total Average Started with any Panel Expansion IR CTR Duration Played Played Expansion Duration Rate Expansion Duration (Seconds) Rate Rate Rate (Seconds) Rate (Seconds) Standard Banner — — 0.14% — — — — — — Rich Media — Commercial Break 4% 2.30 3.96% 3.97 72% 48% 29% — — — Expandable Banner 35% 3.27 0.66% 44.12 25% 62% 48% 41% 17% 52.60 Floating Ad 13% 2.68 4.16% 8.90 65% 51% 30% — — — Formats Floating Ad With 6% 2.38 4.66% — — — — — — — Reminder Floating Expandable 20% 2.58 0.48% — — — — 56% 14% 41.95 In Game 19% — 5.80% 17.17 99% 93% 80% — — — Polite Banner 7% 2.95 0.21% 44.42 59% 67% 51% — — — Push Down Banner 16% 3.46 0.15% 44.72 55% 84% 78% 22% 12% 57.37 Video Strip 23% 4.87 0.18% 63.20 10% 74% 60% 43% 8% 72.39 Interaction Impressions Avg. CTR CTR Avg. Video 50% Fully Total Average Started with any Panel Expansion IR (Rich (Standard Duration Played Played Expansion Duration Rate Expansion Duration Media) Media) (Seconds) Rate Rate Rate (Seconds) Rate (Seconds) Apparel 15% 3.05 0.38% 0.17% 51.92 35% 57% 44% 47% 16% 47.41 Auto 14% 2.85 0.41% 0.15% 49.68 51% 68% 51% 42% 16% 55.11 B2B 9% 3.01 0.44% 0.34% — — — — 12% 6% 46.55 Careers 11% 3.00 0.54% 0.15% — — — — 25% 11% 51.31 Consumer 25% 3.58 0.57% 0.12% 39.68 31% 56% 42% 47% 18% 63.97 Packaged Goods Corporate 18% 3.11 0.32% 0.08% — — — — 59% 26% 74.74 Electronics 21% 2.90 0.35% 0.13% 46.15 36% 72% 59% 42% 17% 48.40 Entertainment 28% 3.35 0.57% 0.18% 54.69 30% 62% 46% 40% 18% 39.44 Financial 17% 3.03 0.30% 0.14% 33.11 45% 62% 43% 36% 17% 64.06 Verticals Gaming 19% 3.48 0.68% 0.21% 74.16 39% 61% 41% 41% 17% 51.12 Government/utilities 20% 3.33 0.47% 0.08% 49.45 41% 64% 42% 41% 16% 51.84 health/Beauty 18% 2.95 0.42% 0.10% 40.92 38% 62% 45% 30% 15% 47.21 Medical 12% 2.92 0.34% 0.28% 52.04 32% 50% 32% 54% 18% 57.12 news/Media 8% 3.04 0.35% 0.15% 93.30 45% 60% 42% 42% 14% 30.76 Restaurant 23% 3.13 0.38% 0.07% 25.33 53% 46% 37% 44% 15% 57.49 Retail 21% 2.83 0.47% 0.15% 42.97 57% 63% 44% 54% 22% 57.30 Services 21% 3.13 0.44% 0.08% 56.89 29% 62% 42% 42% 18% 55.09 Sports 8% 3.05 0.31% 0.10% 32.58 42% 75% 53% 38% 24% 28.36 Tech/Internet 15% 3.05 0.38% 0.14% 32.89 43% 82% 72% 37% 14% 46.39 Telecom 22% 2.98 0.43% 0.13% 49.42 48% 56% 41% 42% 17% 51.14 Travel 22% 3.15 0.49% 0.10% 42.38 59% 69% 57% 40% 14% 46.39 Europe: including Austria, Belgium, Denmark, France, Germany, Greece, Ireland, Italy, netherlands, Poland, Portugal, Spain, Sweden, Switzerland, uK. The benchmark is for a rolling year, last updated Q4 2008. Interaction Rate (IR) user Initiated Interactions' divided by 'Served Impressions' Click-through Rate (CTR) Frequency of Click-throughs as a percentage of served impressions. 'Clicks' divided by 'Impressions'. Avg. Video Duration The average duration the video was played, including user and auto initiated videos (in seconds). Started Rate The number of times the video started out of video's served Impressions 50% Played Rate The number of videos that played over 50% of their total video length divided by 'Video Started'. Fully Played Rate The number of videos that were fully played divided by video that started Total Expansion Rate Total expansions divided by served impressions. Including Auto-initiated expansions Impressions with any Panel Expansion Rate The number of impressions with at least one panel expansion. Avg. Expansion Duration The average time a panel was expanded, including user and auto initiated expansions Rich Media Ads All Eyeblaster's format excluding standard banner, wallpaper and window ads — not Available - either this measurement is not applicable to the format or there was not enough data to be statistically relevant 16 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Appendix 3 Benchmark for Spain Q4 ’07 - Q3 ’08 Performance Metrics (Format and Verticals) Basic Metrics Video Metrics Expandable Metrics Avg. Avg. Video 50% Fully Total Impressions Started Expansion IR CTR Duration Played Played Expansion with any Panel Rate Duration (Seconds) Rate Rate Rate Expansion Rate (Seconds) Standard Banner — 0.18% — — — — — — — Rich Media Commercial Break 4% 3.87% — — — — — — — Formats Expandable Banner 53% 1.02% 73.23 20% 56% 44% 64% 23% 67.38 Floating Ad 8% 2.08% — — — — — — — Polite Banner 6% 0.25% 60.29 47% 65% 50% — — — Push Down Banner 12% 0.21% — — — — — — — Avg. CTR CTR Avg. Video 50% Fully Total Impressions Started Expansion IR (Rich (Standard Duration Played Played Expansion with any Panel Rate Duration Media) Media) (Seconds) Rate Rate Rate Expansion Rate (Seconds) Apparel 21% 0.62% 0.45% 26.86 41% 55% 40% — — — Auto 19% 0.44% 0.18% 21.57 54% 54% 40% 76% 30% 58.70 Consumer Packaged 42% 0.95% 0.15% 31.62 25% 62% 38% — — — Goods Electronics 17% 0.45% 0.16% 40.81 61% 74% 60% 47% 22% 57.02 Entertainment 51% 1.06% 0.22% 35.28 44% 66% 51% 53% 10% 51.19 Financial 14% 0.34% 0.12% 87.28 60% 84% 52% — — — Verticals Gaming 40% 0.94% 0.58% 38.05 44% 62% 45% — — — Government/utilities 13% 0.30% 0.08% — — — — 46% 7% 73.58 health/Beauty 38% 0.81% 0.22% 54.38 35% 62% 46% 43% 13% 51.90 Retail 18% 0.65% — Services 41% 0.54% 0.14% — — — — 79% 23% 63.60 Tech/Internet 9% 0.48% 0.14% 44.00 31% 69% 40% 68% 24% 57.03 Telecom 34% 0.50% 0.39% — — — — — — — Travel 22% 0.68% — 35.64 60% 43% 29% 50% 8% 38.69 The benchmark is for a rolling year, last updated Q4 2008. Interaction Rate (IR) user Initiated Interactions' divided by 'Served Impressions' Click-through Rate (CTR) Frequency of Click-throughs as a percentage of served impressions. 'Clicks' divided by 'Impressions'. Avg. Video Duration The average duration the video was played, including user and auto initiated videos (in seconds). Started Rate The number of times the video started out of video's served Impressions 50% Played Rate The number of videos that played over 50% of their total video length divided by 'Video Started'. Fully Played Rate The number of videos that were fully played divided by video that started Total Expansion Rate Total expansions divided by served impressions. Including Auto-initiated expansions Impressions with any Panel Expansion Rate The number of impressions with at least one panel expansion. Avg. Expansion Duration The average time a panel was expanded, including user and auto initiated expansions Rich Media Ads All Eyeblaster's format excluding standard banner, wallpaper and window ads — not Available - either this measurement is not applicable to the format or there was not enough data to be statistically relevant 17 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Appendix 4 Benchmark for APAC Q4 ’07 - Q3 ’08 Performance Metrics (Format and Verticals) Basic Metrics Video Metrics Expandable Metrics Interaction Impressions Avg. Avg. Video 50% Fully Total Average Started with any Panel Expansion IR CTR Duration Played Played Expansion Duration Rate Expansion Duration (Seconds) Rate Rate Rate (Seconds) Rate (Seconds) Standard Banner — — 0.11% — — — — — — — Rich Media Commercial 3% 2.70 3.14% — — — — — — — Break Expandable 21% 3.32 0.34% 22.96 6% 47% 36% 24% 11% 57.25 Banner Formats Floating Ad 8% 3.00 2.28% 50.89 76% 84% 78% — — — Floating Ad With 5% 2.64 1.60% 11.04 63% 64% 21% — — — Reminder Floating 32% 2.93 0.43% — — — — 78% 32% 151.76 Expandable Polite Banner 2% 2.82 0.11% 40.54 40% 66% 50% — — — Push Down 26% 2.89 0.32% 17.32 23% 63% 47% 45% 11% 24.57 Banner Interaction Impressions Avg. CTR Avg. Video 50% Fully Total Average CTR (Rich Started with any Panel Expansion IR (Standard Duration Played Played Expansion Duration Media) Rate Expansion Duration Media) (Seconds) Rate Rate Rate (Seconds) Rate (Seconds) Apparel 10% 3.11 0.36% 0.18% 17.65 17% 49% 34% 26% 12% 67.31 Auto 7% 2.92 0.23% 0.22% 46.30 27% 78% 69% 17% 10% 52.91 Careers 24% 3.45 0.46% 0.04% — — — — 62% 22% 36.10 Consumer 12% 3.27 0.33% 0.12% 24.46 25% 70% 56% 26% 13% 50.61 Packaged Goods Corporate 5% 2.55 0.12% 0.05% — — — — 10% 5% 50.69 Electronics 16% 2.88 0.32% 0.11% 34.20 34% 64% 50% 29% 15% 56.57 Entertainment 11% 3.09 0.35% 0.16% 29.23 16% 68% 55% 21% 10% 56.96 Financial 8% 2.76 0.15% 0.05% 19.87 22% 64% 52% 30% 8% 74.25 Gaming 6% 4.60 0.30% 0.10% 33.03 20% 54% 41% 6% 3% 46.98 Verticals Government/ 18% 3.77 0.31% 0.18% 69.27 20% 38% 24% 25% 13% 55.22 utilities health/Beauty 12% 2.86 0.45% 0.11% 26.73 18% 47% 34% 23% 13% 62.54 Medical 15% 2.75 0.41% 0.09% — — — — 32% 16% 55.26 news/Media 11% 3.51 0.21% 0.18% — — — — 19% 11% 68.97 Restaurant 27% 4.16 0.67% 0.17% — — — — 37% 17% 32.42 Retail 10% 3.70 0.22% 0.11% 35.47 21% 50% 37% 21% 11% 36.07 Services 9% 2.77 0.14% 0.06% 50.42 13% 51% 35% 15% 8% 53.70 Sports 3% 2.36 0.40% 0.06% — — — — — — — Tech/Internet 4% 2.61 0.12% 0.10% 27.48 27% 58% 38% 23% 11% 67.82 Telecom 13% 3.23 0.31% 0.09% 32.26 11% 52% 40% 34% 15% 49.93 Travel 15% 2.95 0.32% 0.15% 12.99 17% 63% 51% 25% 10% 48.51 APAC: including Australia, China, hong Kong, India, Japan, Korea, Malaysia, new Zealand, Pakistan, Singapore. The benchmark is for a rolling year, last updated Q4 2008. Interaction Rate (IR) user Initiated Interactions' divided by 'Served Impressions' Click-through Rate (CTR) Frequency of Click-throughs as a percentage of served impressions. 'Clicks' divided by 'Impressions'. Avg. Video Duration The average duration the video was played, including user and auto initiated videos (in seconds). Started Rate The number of times the video started out of video's served Impressions 50% Played Rate The number of videos that played over 50% of their total video length divided by 'Video Started'. Fully Played Rate The number of videos that were fully played divided by video that started Total Expansion Rate Total expansions divided by served impressions. Including Auto-initiated expansions Impressions with any Panel Expansion Rate The number of impressions with at least one panel expansion. Avg. Expansion Duration The average time a panel was expanded, including user and auto initiated expansions Rich Media Ads All Eyeblaster's format excluding standard banner, wallpaper and window ads — not Available - either this measurement is not applicable to the format or there was not enough data to be statistically relevant 18 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • Appendix 5 Benchmark for Taiwan Q4 ’07 - Q3 ’08 Performance Metrics (Format and Verticals) Basic Metrics Video Metrics Expandable Metrics Avg. Avg. Video 50% Fully Total Impressions Started Expansion IR CTR Duration Played Played Expansion with any Panel Rate Duration (Seconds) Rate Rate Rate Expansion Rate (Seconds) Standard Banner — 0.04% — — — — — — — Rich Media Expandable Banner 45% 0.58% 307.36 40% 72% 63% 33% 16% 73.48 Formats Floating Ad With 4% 1.97% — — — — — — — Reminder Polite Banner 2% 0.13% 69.28 61% 53% 38% — — — Push Down Banner 2% 0.20% 17.32 23% 63% 47% 29% 0.14% 18.58 Avg. CTR CTR Avg. Video 50% Fully Total Impressions Started Expansion IR (Rich (Standard Duration Played Played Expansion with any Panel Rate Duration Media) Media) (Seconds) Rate Rate Rate Expansion Rate (Seconds) Apparel 6% 0.25% — 58.71 30% 67% 51% Auto 29% 0.54% — 20.32 19% 61% 47% 36% 11% 14.64 Consumer Packaged 5% 0.20% — 72.24 39% 60% 47% — — — Goods Verticals Entertainment 14% 0.35% — 229.87 45% 69% 58% 28% 13% 80.67 Financial 52% 0.74% — — — — — — — — health/Beauty 28% 0.38% — 296.51 42% 71% 62% 26% 17% 85.24 Tech/Internet 2% 0.20% 0.04% — — — — — — — Telecom 36% 0.47% — 93.79 29% 59% 43% 43% 28% 84.96 Travel 18% 0.42% — — — — — 28% 10% 27.70 The benchmark is for a rolling year, last updated Q4 2008. Interaction Rate (IR) user Initiated Interactions' divided by 'Served Impressions' Click-through Rate (CTR) Frequency of Click-throughs as a percentage of served impressions. 'Clicks' divided by 'Impressions'. Avg. Video Duration The average duration the video was played, including user and auto initiated videos (in seconds). Started Rate The number of times the video started out of video's served Impressions 50% Played Rate The number of videos that played over 50% of their total video length divided by 'Video Started'. Fully Played Rate The number of videos that were fully played divided by video that started Total Expansion Rate Total expansions divided by served impressions. Including Auto-initiated expansions Impressions with any Panel Expansion Rate The number of impressions with at least one panel expansion. Avg. Expansion Duration The average time a panel was expanded, including user and auto initiated expansions Rich Media Ads All Eyeblaster's format excluding standard banner, wallpaper and window ads — not Available - either this measurement is not applicable to the format or there was not enough data to be statistically relevant 19 AnALYTICS BuLLETIn ISSuE 3 | MARCh 2009
    • www.eyeblaster.com